Chemistry of Furan Conversion into Aromatics and Olefins over HZSM

On the Way to Biofuels from Furan: Discriminating Diels–Alder and Ring-Opening Mechanisms. S. Vaitheeswaran , Sara K. Green , Paul Dauenhauer , and ...
0 downloads 0 Views 5MB Size
RESEARCH ARTICLE pubs.acs.org/acscatalysis

Chemistry of Furan Conversion into Aromatics and Olefins over HZSM-5: A Model Biomass Conversion Reaction Yu-Ting Cheng and George W. Huber* Department of Chemical Engineering, 159 Goessmann Lab, University of MassachusettsAmherst, 686 North Pleasant Street, Amherst, Massachusetts 01003-9303, United States ABSTRACT: The conversion of furan (a model of cellulosic biomass) over HZSM-5 was investigated in a thermogravimetric analysismass spectrometry system, in situ Fourier transform infrared analysis, and in a continuous-flow fixed-bed reactor. Furan adsorbed as oligomers at room temperature with a 1.73 of adsorbed furan/Al ratio. These oligomers were polycyclic aromatic compounds that were converted to CO, CO2, aromatics, and olefins at temperatures from 400 to 600 °C. Aromatics (e.g., benzene, toluene, and naphthalene), oligomer isomers (e.g., benzofuran, 2,2-methylenebisfuran, and benzodioxane), and heavy oxygenates (C12þ oligomers) were identified as intermediates formed inside HZSM-5 at different reaction temperatures. During furan conversion, graphite-type coke formed on the catalyst surface, which caused the aromatics and olefins formation to deactivate within the first 30 min of time on-stream. We have measured the effects of space velocity and temperature for furan conversion to help us understand the chemistry of biomass conversion inside zeolite catalysts. The major products for furan conversion included CO, CO2, allene, C2C6 olefins, benzene, toluene, styrene, benzofuran, indene, and naphthalene. The aromatics (benzene and toluene) and olefins (ethylene and propylene) selectivity decreased with increasing space velocity. Unsaturated hydrocarbons such as allene, cyclopentadiene, and aromatics selectivity increased with increasing space velocity. The product distribution was selective to olefins and CO at high temperatures (650 °C) but was selective to aromatics (benzene and toluene) at intermediate temperatures (450600 °C). At low temperatures (450 °C), benzofuran and coke contributed 60% of the carbon selectivity. Several different reactions were occurring for furan conversion over zeolites. Some important reactions that we have identified in this study include DielsAlder condensation (e.g., two furans form benzofuran and water), decarbonylation (e.g., furan forms CO and allene), oligomerization (allene forms olefins and aromatics plus hydrogen), and alkylation (e.g., furan plus olefins). The product distribution was far from thermodynamic equilibrium. KEYWORDS: furan, catalytic fast pyrolysis, biomass, aromatics, olefins, zeolites, HZSM-5, hydrocarbon pool

1. INTRODUCTION Due to its low cost and availability, lignocellulosic biomass is receiving significant attention worldwide as a feedstock for renewable liquid fuels.14 Lignocellulosic biomass is not currently used as a feedstock to make liquid fuels due to technological and economic challenges.1,4 Several different processes for obtaining biofuels from biomass are currently under development.5,6 The ideal process to produce biofuels from lignocellulosic biomass would be a single step reactor at short residence times where solid biomass is directly converted into a liquid fuel. Catalytic fast pyrolysis (CFP) is a promising process that occurs in a single reactor at short residence times with solid biomass being directly converted into gasoline-range aromatics.79 In CFP, solid biomass is first pyrolyzed in the presence of a zeolite catalyst. The biomass is rapidly heated (>500 °C/s) to intermediate temperatures (400600 °C) in the first step. The importance of pyrolysis heating rate is well-known.10,11 The biomass feedstock decomposes into pyrolysis vapors at these temperatures. The pyrolysis vapors then enter the zeolite pores, in which they are converted into aromatics, CO, CO2, H2O, and dehydrated oxygenates (furan and furfural).7,1018 Biomass consists of hydrogen-deficient compounds.19,20 The goal of biomass conversion processes is to reject oxygen as a r 2011 American Chemical Society

combination of CO, CO2, and water and produce hydrocarbon products that have an enriched hydrogen and lower oxygen content. An effective hydrogen-to-carbon ratio (H/Ceff) can be used to help explain the hydrogen deficiency of biomass. The H/ Ceff is defined as eq 1, where H, C, and O are the moles of hydrogen, carbon, and oxygen, respectively.19,20 Our research group has shown that the product yield from zeolite conversion of different biomass feedstocks is a function of the H/Ceff ratio, with higher yields being obtained with feedstocks of higher H/ Ceff ratio.21 H  2O ð1Þ C The H/Ceff ratio of biomass-derived oxygenates is lower than that of petroleum-derived feedstocks due to the high oxygen content in biomass-derived compounds. The H/Ceff ratio of petroleum-derived feeds ranges from ∼2 for highly parafinic feeds to slightly greater than 1 for feeds with high aromatic content. The H/Ceff ratio of biomass-derived feedstock such as cellulose and glycerol are 0 and 0.67, respectively. H=Ceff ¼

Received: February 24, 2011 Revised: April 18, 2011 Published: April 26, 2011 611

dx.doi.org/10.1021/cs200103j | ACS Catal. 2011, 1, 611–628

ACS Catalysis Zeolites offer one potential method for the conversion of hydrogen-poor biomass into fuels and chemicals. Researchers have used zeolites to directly convert biomass-derived carbohydrates into aromatics and olefins, beginning first in, the early 1980s with the work of Chen et al.20,22 Chen et al. fed sugars such as glucose and xylose, over HZSM-5 catalyst and produced aromatics in low yields (40%) oxygen contents. The tar can be formed from furan-like intermediates. Increasing the H/Ceff ratio in feeds by adding methanol, or increasing reaction temperatures can reduce tar formation and increase aromatic yields. Horne and Williams also used HZSM-5 to catalytically pyrolyze woods, bio-oils, and vapors in the 1990s.2830 In their study, furans and other oxygenates such as phenols were found in the liquid products, suggesting that these furans are important reaction intermediates during zeolite deoxygenation. Resasco et al. have used HZSM-5 catalyst to convert propanal, a “representative biomass-derived oxygenate”, into aromatics.3133 The conversion pathway involves aldol condensation to form propanal dimer and trimer.31 They also demonstrated that the addition of mesopores into HZSM-5 catalyst was able to overcome diffusion limitation and coking reaction for propanal conversion, which are major concerns for biomass conversion over zeolites.32 We have previously shown that both furfural and furan were major intermediates from CFP of glucose and cellulose at 600 °C.7,12 These products can account for up to 40% of the carbon in the oxygenated products.7,12 Furfural, furan, glucose, and cellulose all had similar aromatic product distribution for CFP.34 This indicates that furan is most likely an important reaction intermediate for the CFP of biomass. In addition, both furan and furfural are thermally stable molecules that do not decompose in the gas phase, unlike glucose or cellulose. Furan can only be thermally decomposed into CO and allene with low yield ( CO2 ∼ xylenes. Intermediates that were identified in order of decreasing selectivity include coke . indene > naphthalene ∼ cyclopentadiene > styrene ∼ benzofuran > allene ∼ methylfuran ∼ methylindene > methylnaphthalene ∼ indane > ethylbenzene ∼ hexatriene ∼ hexadienyne∼ furylethylene ∼ methylstyrene ∼ dihydronapthlane. The coke selectivity decreased with increasing furan conversion. The “coke” may include compounds retained inside HZSM-5 pores, including intermediates such as benzofuran and indene. These results were consistent with results of TGAMS in which furan adsorbed as oligomers on HZSM-5. 616

dx.doi.org/10.1021/cs200103j |ACS Catal. 2011, 1, 611–628

ACS Catalysis

RESEARCH ARTICLE

Table 2. Carbon Yield (%), Deactivation Rate, Catalyst Activity, and wt % of Coke on Catalyst As a Function of Time on-Streama time on-stream (min) 00.5

2.53

8.59

27.528

59.560

119.5120

furan

23.40

47.97

68.42

81.74

85.50

78.73

CO

7.00

5.78

3.44

0.00

0.00

0.00

CO2

1.24

0.39

0.10

0.00

0.00

0.00

C2H4 (ethylene)

3.54

2.79

1.24

0.23

0.09

0.00

C3H6 (propylene)

2.74

2.47

1.75

0.15

0.03

0.00

C4 olefins

0.37

0.34

0.47

0.03

0.00

0.00

C6H6 (benzene)

4.62

5.19

3.98

0.12

0.05

0.00

C7H8 (toluene) C8H10 (ethylbenzene)

3.55 0.21

3.40 0.44

2.18 0.25

0.14 0.06

0.08 0.00

0.00 0.00

C8H10 (xylenes)

0.71

0.81

0.21

0.09

0.00

0.00

C8H8 (styrene)

0.76

0.48

0.19

0.05

0.00

0.00

C8H6O (benzofuran)

0.14

0.13

0.08

0.03

0.02

0.00

C9H8 (indene)

0.65

0.49

0.26

0.09

0.07

0.03

C10H8 (naphthalene)

0.41

0.17

0.09

0.08

0.07

0.00

C3H4 (allene)

0.31

0.33

0.39

0.21

0.10

0.00

C5H6 (cyclopentadiene) C5H6O (methylfuran)

0.60 0.07

1.24 0.38

0.76 0.42

0.04 0.05

0.00 0.02

0.00 0.00

C6H8 (hexatrienes)

0.19

0.11

0.08

0.00

0.00

0.00

C6H6 (hexadienyne)

0.05

0.10

0.08

0.04

0.00

0.00

C6H6O (furylethylene)

0.00

0.06

0.02

0.00

0.00

0.00

C9H12 (trimethylbenzene)

0.05

0.06

0.06

0.12

0.00

0.05

C9H10 (methylstyrene)

0.06

0.08

0.03

0.00

0.00

0.00

C9H10 (Indane)

0.09

0.09

0.05

0.00

0.00

0.00

C10H10 (methylindene) C10H10 (dihydronaphthalene)

0.24 0.00

0.13 0.00

0.09 0.00

0.06 0.00

0.00 0.00

0.00 0.00

C11H10 (methylnaphthalene)

0.13

0.15

0.02

0.02

0.02

0.00

coke (est)

48.88

26.44

15.36

16.65

13.95

21.20

CO þ CO2

8.24

6.17

3.53

0.00

0.00

0.00

olefins

7.79

7.37

4.76

0.70

0.22

0.00

11.62

11.61

7.48

0.85

0.31

0.08

aromatics deactivation rate (eq 4)

126.0

catalyst activity (eq 5)

51.2

23.4

5.5

9.5

0.68

0.41

0.24

0.19

0.28

11.02

22.25

60.77

115.11

269.96

total coke on catalyst (wt coke/wt cat.) a

2.98

Reaction conditions: temperature, 600 °C; furan partial pressure, 6 Torr; and WHSV, 10.4 h1.

not change with increasing furan conversion. However, the CO þ CO2 and olefins increased with increasing furan conversion. This further suggests that the “coke” was larger hydrocarbons that were trapped inside the zeolite at low conversion. 3.5. Effect of Temperature. Table 4 shows the product selectivity for furan conversion as a function of temperature. At 450 °C, benzofuran was the primary product observed.37 The carbon selectivity to benzofuran decreased from 17.7 to 1.7% as the temperature increased from 450 to 650 °C. As the temperature increased, the major products became CO, ethylene, benzene, propylene and toluene. At the higher temperature, the selectivity to highly reactive species, including C4 olefins, allene, cyclopentaidene, and hexadienyne increased. The selectivity of xylenes, ethylbenzene, methylfuran, furylethylene, and methylindene slightly decreased with increasing temperature. Carbon

This suggests that the type of coke that is formed inside the zeolite changed with furan conversion. The indane and indenes were most likely interconverted through a dehydrogenation reaction. The indenes were converted into aromatics, olefins, and coke, as will be shown later in this paper. Benzofuran was formed by DielsAlder reaction of two furan molecules. Benzofuran was further converted into CO, aromatics, and coke, as will be shown later in this paper.37 Naphthalene and methylnaphthlene most likely formed coke and hydrogen. In Table 3, we also grouped the products into CO þ CO2, olefins (including ethylene, propylene, C4 olefins, allene, cyclopentadiene, hexatriene, and hexadienyne) and aromatics (including benzene, toluene, ethylbenzene, xylene, styrene, benzofuran, indene, naphthalene, methylstyrene, indane, dihydronaphthalene, and methylnaphthalene). The carbon selectivity of aromatics did 617

dx.doi.org/10.1021/cs200103j |ACS Catal. 2011, 1, 611–628

ACS Catalysis

RESEARCH ARTICLE

Table 3. Carbon Product Selectivity As a Function of Space Velocitya WHSV (h1)

21.86

10.35

1.95

furan conversion

0.28

0.48

0.97

carbon deposited on active sites (mol/mol)

8.97

6.27

1.77

carbon yield of coke (%)

14.91

14.44

21.56

products

a

carbon selectivity (%)

CO

11.46

13.91

product

CO2

0.17

1.13

3.30

product

C2H4 (ethylene) C3H6 (propylene)

6.00 4.99

7.40 6.71

11.10 7.96

product product

C4 olefins

0.73

0.83

0.79

C6H6 (benzene)

7.10

8.03

11.60

product

C7H8 (toluene)

6.38

7.32

10.69

product

C8H10 (ethylbenzene)

0.40

0.37

0.11

intermediate

C8H10 (xylenes)

1.43

1.35

2.72

product

C8H8 (styrene)

2.68

2.53

1.42

intermediate

C8H6O (benzofuran) C9H8 (indene)

2.19 5.19

1.84 4.13

0.25 2.33

intermediate intermediate

C10H8 (naphthalene)

3.31

2.16

1.14

intermediate

C3H4 (allene)

1.39

0.88

0.16

intermediate

C5H6 (cyclopentadiene)

3.12

2.71

0.50

intermediate

C5H6O (methylfuran)

1.55

0.86

0.00

intermediate

C6H8 (hexatrienes)

0.47

0.32

0.00

intermediate

C6H6 (hexadienyne)

0.39

0.26

0.00

intermediate

C6H6O (furylethylene) C9H12 (trimethylbenzene)

0.26 0.00

0.16 0.02

0.00 0.00

intermediate unknown

C9H10 (methylstyrene)

0.37

0.32

0.14

intermediate

C9H10 (Indane)

0.57

0.49

0.20

intermediate

C10H10 (methylindene)

1.59

1.36

0.43

intermediate

0.12

0.18

0.00

intermediate

C11H10 (methylnaphthalene)

0.77

0.94

0.37

intermediate

33.80 15.04

27.05 21.03

olefins

17.08

19.11

20.50

aromatics

32.09

31.01

31.42

450

intermediate

C10H10 (dihydronaphthalene)

37.39 11.64

reaction temp (°C)

role

17.72

coke CO þ CO2

Table 4. Carbon Product Selectivity As a Function of Reaction Temperaturea

0.22 6.7

11.5

13.9

17.9

CO2 C2H4 (ethylene)

0.0 3.9

0.0 4.3

1.1 7.4

1.1 11.4

0.60

C3H6 (propylene)

2.3

3.8

6.7

7.9

C4 olefins

0.1

0.5

0.8

1.4

C6H6 (benzene)

3.6

4.9

8.0

9.4

C7H8 (toluene)

4.2

5.4

7.3

6.6

C8H10 (ethylbenzene)

0.7

0.6

0.4

0.2

C8H10 (xylenes)

1.5

1.3

1.3

0.9

0.8 17.7

1.2 4.4

2.5 1.8

1.9 1.7 3.0

C9H8 (indene)

3.4

3.0

4.1

C10H8 (naphthalene)

1.8

3.0

2.2

1.2

C3H4 (allene)

0.0

0.1

0.9

2.0

C5H6 (cyclopentadiene)

0.1

0.6

2.7

3.7

C5H6O (methylfuran)

1.4

1.3

0.9

0.5

C6H8 (hexatrienes)

0.0

0.4

0.3

0.1

C6H6 (hexadienyne) C6H6O (furylethylene)

0.0 0.3

0.0 0.3

0.3 0.2

0.3 0.1

C9H12 (trimethylbenzene)

0.0

0.1

0.0

0.0

C9H10 (methylstyrene)

0.2

0.2

0.3

0.6

C9H10 (Indane)

1.1

1.6

0.5

0.2

C10H10 (methylindene)

2.0

1.3

1.4

0.8

C10H10 (dihydronaphthalene)

0.0

0.0

0.2

0.1

C11H10 (methylnaphthalene)

0.7

0.5

0.9

0.4

CO þ CO2 olefins aromatics

Reaction conditions: 600 °C; furan partial pressure, 6 Torr.

0.48

650

CO

coke

a

0.32

600

furan conversion

C8H8 (styrene) C8H6O (benzofuran)

intermediate

500

47.4

49.7

33.8

26.9

6.7

11.5

15.0

19.0

6.4

9.7

19.1

26.7

37.7

27.3

31.0

26.8

Reaction conditions: WHSV, 10.4 h1; furan partial pressure, 6 Torr.

fed these reactants using a helium carrier gas (408 mL/min) through a bubbler containing benzofuran or indene. The reaction temperature was 600 °C for each run. Table 5 shows the carbon selectivity of products obtained from benzofuran and indene conversion. The major products from benzofuran conversion were CO, benzene, toluene, and coke. These products accounted for 91% of the products. Benzene was the primary aromatic being produced in 41% carbon selectivity. Only trace amounts of olefins were observed ( styrene > naphthalene > xylenes > methylindene > methylnaphthalene > ethylbenzene. These results indicate that olefins did not form from benzofuran, but were formed from a separate pathway. This also suggests that during furan conversion, olefins and aromatics formed by different reaction pathways. If all the oxygen is removed by decarbonylation, then the CO selectivity should be 12.5%. In our study, we obtained 87% of this value (Table 5, 10.87 of 12.5). This suggests that benzofuran underwent decarbonylation reactions.

dioxide was observed only at temperatures of 600 and 650 °C. The styrene selectivity went through a slight maximum of 2.5% carbon selectivity at 600 °C. The indene selectivity varied between 3.0 and 4.1% with temperature. The naphthalene selectivity varied between 1.2 and 3.0% with no systematic trend. The coke selectivity decreased with increasing temperature. The olefins and CO þ CO2 selectivity increased with increasing temperature. The selectivity of olefins was less than aromatics at 450600 °C, but was comparable to aromatics at 650 °C. This indicates that olefins were favorable at high temperatures, but aromatics were favorable at intermediate temperatures. Similar to the WHSV effect, more olefins, CO, and CO2 were extracted from the retained intermediates at high furan conversion. 3.6. Conversion of Benzofuran and Indene. Benzofuran and indene were identified as two important intermediates. We also fed them into the flow fixed-bed reactor (benzofuran, 99%; indene, 98%; both from Sigma-Aldrich). Their vapor pressures are quite low, making it impossible to feed by a syringe pump. We 618

dx.doi.org/10.1021/cs200103j |ACS Catal. 2011, 1, 611–628

ACS Catalysis

RESEARCH ARTICLE

shows the GPC results of the leached solution, showing that the molecular weight of the retained compounds can be as high as 106 (this is explained later in this paper). The concentration of species observed by GPC decreased with increasing temperature, especially for the heavy compounds (short retention time). The negative peak at a Mw of 100 g/mol was caused by DCM, which is shown in Figure 7, where pure DCM has a negative peak at the retention time equal to 100 g/mol. As shown in Figure 8, when increasing temperature, the identified Mw distribution moved from high Mw to low Mw (excluding the negative peak area from the DCM solvent). One explanation for this phenomena is that the retained compounds polymerized with increasing temperature (probably in polycyclic aromatic structures), and we were not able to analyze these polymers, since they cannot be dissolved in DCM. We did see a lot of blackish particles suspended and aggregated in both HF and DCM phases at elevated temperatures, which we were not able to analyze with any technique. At 300 and 400 °C, the furan polymers began to decompose, and oxygen was removed as CO and CO2. Those retained compounds formed products and unknown intermediates that were insoluble in DCM due to their structure and high molecular weight. At 500 and 600 °C, the furan polymers were converted into gaseous products with only small amounts of soluble products detected in the DCM, as shown in Figure 6. These retained compounds at these temperatures were graphite-type coke that were insoluble in DCM. Among the retained compounds shown in Tables 6 and 7, furan showed high selectivity at room temperature (42%) and gradually decreased (9.2%) as the reaction temperature increased, as shown in Table 6. Furan and the other oxygenates made up to 80% of the identified carbon by GC/MS at most temperatures. This indicates that the oxygenates were the major species inside the zeolite catalyst. Deoxygenated aromatics (benzene, toluene, naphthalene, and methylnaphthalene) were formed inside the HZSM-5, usually at temperatures higher than 400 °C. They were observed only in trace quantities at temperatures below 400 °C. Benzene and toluene were final products that have been shown in previous results. Benzofuran and dibenzofuran (19) were formed via DielsAlder condensation. Benzodioxane (13) and 2,2-methylenebisfuran (6) showed the highest selectivity at 300 and 100 °C, respectively. To understand the formation of these compounds will be a future work. The number of retained carbon per aluminum site was also calculated as shown in Table 6. It decreased with increasing temperature, indicating that less mass was trapped inside HZSM5 at higher temperatures. In addition to compounds retained inside HZSM-5, we found that there were tars condensed at the outlet of the quartz reactor after reactions at 400600 °C. Their composition was identified by GC/MS and is shown in Table 6 (compounds 11, 14, 19, and 2731). Most of them were deoxygenated and were polycyclic aromatics. Their formation was probably due to crackings from heavy intermediates or continued alkylation or dealkylation from aromatics. The UVvis spectra of the DCM solutions from leaching of the zeolitefuran mixtures are shown in Figure 9. UVvis is a good technique to identify conjugated double bonds.42,50 At room temperature, strong absorption bands were observed at around 234, 274, and 364 nm. These bands were formed from compounds that have no more than six conjugated double bonds, indicating that they were smaller in size than furan trimers.50 A red shift toward more delocalized conjugated double bonds was observed as the temperature of the furanzeolite mixture

Table 5. Carbon Selectivity (%) of Products Obtained by Using Benzofuran or Indene As the Feedstocka products CO C2H4 (ethylene) C3H6 (propylene)

benzofuran

indene

10.87

0.00

0.45 0.42

5.01 1.25

C4H4O (furan)

1.99

0.00

C6H6 (benzene)

41.01

16.81

C7H8 (toluene)

13.87

9.71

C8H10 (ethylbenzene)

0.13

1.42

C8H10 (xylenes)

0.64

1.04

C8H8 (styrene)

1.52

4.23

C9H8 (indene) C10H8 (naphthalene)

2.81 0.80

x 4.10

C3H4 (allene)

0.00

0.86

C9H10 (indane)

0.00

2.76

C10H10 (methylindene)

0.35

1.62

C11H10 (methylnaphthalene)

0.17

0.00

coke

24.98

51.21

CO þ CO2 olefins

10.87 0.86

0.00 7.11

aromatics

61.30

41.68

Reaction conditions: temperature, 600 °C; helium carrier gas, 408 mL/ min; WHZSM-5, 57 mg; feedstocks were fed by a bubbler. a

Indene produced coke (51% carbon selectivity), aromatics (42% carbon selectivity), and olefins (7% carbon selectivity). Benzene, toluene, and styrene were the primary aromatics produced from indene. Ethylene was the primary olefin produced from indene. It produced only ethylene and did not produce any of the higher olefins. It should be pointed out that the ratio of the moles of benzene to moles of ethylene was 1.1. This suggests that ethylene and benzene are made from a common intermediate (possibly by dealkylation of styrene). Trace quantities of indane and methylindene were formed, most likely from hydrogenation and alkylation of indene, respectively. Other products that were observed include ethylbenzene, xylenes, naphthalene, and allene. High yields of coke were observed (>50% coke selectivity), indicating that indene was a coke precursor. 3.7. Identification of Products Inside the Zeolite during Furan Conversion. As described in Section 2.3, we extracted compounds retained inside HZSM-5 during furan conversion using HF and DCM. The products were obtained by running the flow fixed-bed reactor at various temperatures (25600 °C) at a space velocity (WHSV) of 2.36 h1. The furan partial pressure was 6 Torr. After 4 min time on-stream, the reactor was quenched to room temperature, followed by the leaching process mentioned in Section 2.3. The 1/20 volumetrically diluted solutions were clear and showed various colors. The solutions after leaching of the zeolite were yellowish for the 500 and 600 °C samples and black for all other samples. The compounds identified by GC/MS are shown in Table 6 with their structures drawn as compounds 126 in Table 7. Most of these compounds were oxygenates with conjugated double bonds. However, they contributed to only less than 5% of the overall carbon that was in the zeolite. Volatile organics were lost during the process. In addition, most of the retained compounds were high molecular weight, which cannot be detected using GC. Figure 6 619

dx.doi.org/10.1021/cs200103j |ACS Catal. 2011, 1, 611–628

ACS Catalysis

RESEARCH ARTICLE

Table 6. Carbon Selectivity of Retained Compounds Obtained from HF Leaching Experimenta species

C4H4O

C4H8O

C6H6

C7H8

C8H6O

C9H8O2

no.

1

2

3

4

5

6

7

8

9

RT

42.2

0.9

0.0

0.0

8.1

2.0

0.0

0.0

1.2

0.0

0.0

100

2.3

0.1

0.0

0.0

11.2

14.0

0.0

0.1

0.5

0.0

0.1

0.0

200

3.3

2.3

0.0

0.0

14.7

1.6

0.0

1.3

0.0

0.0

0.7

20.0

300

6.8

2.5

0.4

0.4

4.8

3.3

3.4

1.0

0.0

1.1

0.9

4.6

400

0.1

0.2

0.5

0.0

0.9

2.5

1.1

2.0

0.0

2.8

3.6

2.5

500

7.3

14.6

20.0

10.9

2.6

0.0

0.0

0.0

0.0

0.0

29.7

0.0

600

9.2

48.4

20.2

6.0

0.0

0.0

0.0

0.0

0.0

0.0

13.1

0.0

C11H10O

C10H8O2

species no.

C8H8O2

C11H10

13

14

C11H10O2 15

C8H8O 16

C12H12O2 17

C11H8O 18

C6H6O

C9H8O

C12H8O 19

C11H14O 20

C11H12O2

C10H8O 21

C7H8O 10

C12H10O2 22

C10H8 11

23

C10H10O 12 0.0

24

RT

0.0

0.1

3.3

0.0

0.8

0.0

0.0

0.0

0.0

0.0

0.0

100

0.0

0.1

23.3

0.0

4.5

0.0

0.0

0.0

0.0

2.4

0.0

0.7

200

9.7

0.9

0.0

0.0

5.4

4.0

16.8

0.0

0.0

19.3

0.0

0.0

300

50.3

0.6

0.0

4.3

0.0

0.0

3.7

7.7

4.2

0.0

0.0

0.0

400

3.9

1.7

0.0

1.6

0.0

0.0

0.9

2.3

41.4

0.0

32.0

0.0

500 600

0.0 0.0

7.8 3.1

0.0 0.0

0.0 0.0

0.0 0.0

0.0 0.0

0.0 0.0

0.0 0.0

7.2 0.0

0.0 0.0

0.0 0.0

0.0 0.0

species

C15H14O3

C12H12O3

H/Ceff

% identified C

μmol of C/g. cat.

no.

25

26

RT

14.0

27.2

0.4

5.22

1876

100

32.8

7.8

0.4

4.60

1488

200

0.0

0.0

0.4

0.73

223

300

0.0

0.0

0.5

2.57

605

400 500

0.0 0.0

0.0 0.0

0.4 0.8

2.09 0.23

353 33

600

0.0

0.0

1.1

0.48

49

0.4

Spent catalyst was obtained from furan conversion over HZSM-5 in the flow fixed-bed reactor. Reaction conditions: WHSV, 2.36 h1; furan partial pressure, 6 Torr; reaction time, 4 min (see also Table 7). a

increased from RT to 300 °C which was evidence that the conjugated double bonds grew with increasing temperature inside HZSM-5. The number of conjugated double bonds at 300 °C was much greater than six (close to infinity).42,50 The spectra intensity decreased dramatically at 400 °C. This can be explained by the observation that most compounds were converted to gas phase products and insoluble products at this temperature. However, compounds having conjugated double bonds still existed in the DCM solutions and resulted in absorption under UVvis irradiation. At 500 and 600 °C, the intensity was low, since the produced graphite-type coke was insoluble in DCM. In Table 7, we also found that almost all retained compounds identified by GC/MS have conjugated double bonds. However, although conjugated double bonds were identified by UVvis and GC/MS, it is not possible to infer correct structures of those compounds. In addition to condensed phase products, gas phase products were collected in air bags and analyzed by GC/FID. As shown in Figure 10, furan conversion went through a minimum at 400 °C and then increased with increasing temperature. The decrease before 400 °C could be attributed to the blocking of zeolite pores by furan polymers. At 400 °C, gas phase products such as CO, aromatics, and olefins were formed, but the reaction rate was low. At 500 and 600 °C, the reaction routes were shifted to the formation of gas phase products. Compounds that did not diffuse

out formed graphite-type coke. The gas phase products distribution is consistent with TGAMS and in situ FTIR where products can be formed only at temperatures higher than 400 °C. The slightly restored signal in FTIR representing a Brønsted acid site was attributed to the formation of gas phase products. It was not possible to identify all retained compounds in the leaching experiments. Most compounds were volatile or insoluble in DCM. In addition, we lost some mass during the HF dissolution process due to the increasing temperature caused by heat of dissolution (exothermic process). The acidic circumstances also caused the polymerization of intermediates. In another test, we added 1 mL pf furan into 5 mL of HF and HCl. Then we used 5 mL of DCM to extract organic compounds, and the organic samples were subjected to GPC analysis. Figure 7 shows that furan polymerized in HF and HCl. It is interesting that the furan polymers also showed two different Mw distributions in the strong acid solution (HCl). We repeated the leaching/GPC experiment described in Sections 2.3 and 3.7 by using nonporous silica/alumina as a catalyst. The reactions were run at RT and 300 °C. As shown in Figure 7, the furan was again polymerized and had two major Mw regions. Figure 7 suggests that the existence of compounds that have Mw ∼106 probably was due to the limitation of the GPC column (the limitation of the Mw for the GPC column is 104), since no matter how we 620

dx.doi.org/10.1021/cs200103j |ACS Catal. 2011, 1, 611–628

ACS Catalysis

RESEARCH ARTICLE

Table 7. Structures of Identified Retained Compounds (see also Table 6)

Figure 6. Molecular weight distribution of species inside HZSM-5 during furan conversion at various temperatures. Reaction conditions: WHSV, 2.36 h1; furan partial pressure, 6 Torr; reaction time, 4 min.

polymerized furan, it showed Mw ∼106 after ∼104. We concluded that furan formed polymers that have Mw (>104) beyond the limitation of the GPC column. 3.8. Coke Combustion. Figure 11 shows the TPO curves obtained in the combustion of the coke deposited on the catalyst after reactions at different temperatures (25, 300, 400, and 600 °C). Deconvolution of each curve is shown in Table 8 using four different curves with peak temperatures of 523560, 433489, 218, and 50 °C. In Table 8, the oxygenated coke means heavy oxygenates retained inside zeolites (e.g., furan polymers). The peaks at temperatures from 523 to 560 °C most likely represented graphitic coke consisting of large polycyclic aromatics.25,49 The lower-temperature coke was hydrocarbon

species trapped inside the zeolite or on the pores that probably contained oxygen. As the reaction temperature increased, the TPO curves were shifted to a higher temperature. However, the high-temperature coke formed even when furan was adsorbed at room temperature. This coke contributed almost 50% of the entire coke at room temperature. It was surprising that the graphite-type coke formed at 25 °C, since no CO or CO2 were observed in the gas phase. At 600 °C, more than 90% coke was graphite-type coke. As shown in Table 8 the highest-temperature deconvolution peak gradually shifted from 523 to 560 °C, and its area ratio increased as the reaction temperature increased. An area ratio drop at 400 °C (0.62 at 300 °C but 0.48 at 400 °C) was found, which was consistent with TGAMS; FTIR; and the 621

dx.doi.org/10.1021/cs200103j |ACS Catal. 2011, 1, 611–628

ACS Catalysis

RESEARCH ARTICLE

Table 10 shows that the molar carbon ratios of (benzene þ toluene) to (ethylene þ propylene) were 0.831.28 in all reaction conditions. The thermodynamic calculations also show that toluene is more favorable than benzene. However, more benzene than toluene was produced for furan conversion over HZSM-5. ASPEN software was also used to calculate the thermodynamic product distribution of furan at different temperatures, as shown in Table 10. We used the built-in thermodynamic data from ASPEN and the Peng-Robbinson EOS to do these calculations. Table 10 shows that the product distribution was far from thermodynamic equilibrium. For example, experimental data showed that the benzene/toluene ratios were close to 1, benzene was always more than naphthalene (benzene/naphthalene >1), and the ethylene/propylene ratio was only a little bit higher than 1. However, simulation results showed that benzene/toluene and ethylene/propylene ratios for all temperatures were much greater than 1, and naphthalene was more thermodynamic favorable than benzene (benzene/naphthalene butyldiene > hexadienyne > hexatriene in most reaction conditions. Looking at reactions 1216, the olefin selectivity can be used to explain why the aromatic selectivity decreased as toluene > indene > styrene > naphthalene > methylindene. Only reactions at 450 and 500 °C did not follow this rule. For the reaction at WHSV 1.95 h1, although hexadienyne showed higher selectivity than butyldiene, naphthalene showed higher selectivity than styrene, which means it did not break the rule. Alkylation reactions also likely occur during furan conversion. The alkylation reactions are thermodynamically favorable (Table 9, reaction 14). Alkylation reactions have been suggested in the hydrocarbon pool mechanism for biomass model compounds reacting on zeolites.43,6770 For example, benzene can be alkylated by ethylene to form ethylbenzene and styrene (if followed by dehydrogenation) on HZSM-5.71 Chica et al. proposed a mechanism for thiophene conversion over HZSM-5 in which benzothiophene is produced via thiophene alkylation with reactive intermediates (ring-opened thiophene).43 Haw et al. also suggested that during MTO process, alkylation by methanol probably causes chain growth of intermediates.67 Methylacetylene can also be polymerized over zeolites (Y, beta, and HZSM-5) to form methylbenzenes.65 Propylene is also oligomerized to form aromatics over HZSM-5 catalyst, where adsorbed olefins can undergo DielsAlder condensation to form cycle olefins and aromatics.72,73 In this study, we suggest toluene is continuously alkylated by methylacetylene to form styrene, indene, and naphthalene (reactions 1719). Benzene can also be alkylated with propylene to form toluene and xylenes (reactions 20 and 21). More alkylated product, trimethylbenzene, was occasionally observed in our experimental conditions. In addition, alkylation on propylene leads to chain growth and cyclization (reactions 2224). Ethylene has been considered produced during the formation of aromatics and is involved in mechanisms different from other olefins.74 In Table 10, we calculated the molar ratio of ethylene to different aromatics (benzene þ toluene þ xylenes þ styrene þ indene þ naphthalene) at different reaction conditions. The ratios are close to unity, except for reactions at high WHSV (21.86 h1) and high temperature (650 °C). This indicates that the formation of aromatics accompanies the production of ethylene. In the study of propylene recycle, we found that propylene cofeed helped formation of alkylated benzenes.8 This indicates that alkylation is a possible route to form aromatics. To identify correct reactions about aromatics formation will be a future work. An HZSM-5 cage that has a hydrocarbon pool and acidic active site inside is viewed as a “supramolecular structure.” It plays an important role during reaction.75 Biomass feedstocks can be converted to either intermediates that contribute to the hydrocarbon pool or reactants that provide hydrocarbons to react with the produced hydrocarbon pool and form products. However, the requirement to keep the hydrocarbon pool active is that the feedstock itself should be hydrogen-rich, since the hydrocarbon pool is essentially hydrogen-poor; otherwise, the hydrocarbon pools will grow to form heavy hydrocarbons and coke and block the zeolite channel. During furan conversion, for example, the hydrogen-poor intermediate, methylacetylene, is very easily polymerized to form coke.64,65 In addition, furan itself can be condensed on the HZSM-5 surface via the mechanism similar to benzofuran formation and, finally, form coke. Although the blocked cages still are active for

Table 9. Gibbs Free Energy for Furan Conversion into Various Products, Calculated from Data in Ref 51 ΔGrxn reaction 1

2

3

4

5

C4H4O (furan) f C2H4 þ CO þ C

2C4H4O f C3H6 þ 2CO þ H2 þ 3C

2C4H4O f C6H6 þ 2CO þ H2

3C4H4O f C7H8 þ 3CO þ 2H2 þ 2C

3C4H4O f C8H10 þ 3CO þ C þ H2

temp

(kJ/

(°C)

mol)

600

269

500

246

450

234

600

380

500

351

450

336

600

304

500 450

276 262

600

528

500

486

450

465

600

464

500

434

450

418

6

3C4H4O f C9H8 þ 3CO

600 500

419 377

450

355

7

3C4H4O f C10H8 þ 4CO þ 4H2 þ 2C

600

615

500

565

8

9

10

11

12

C4H4O f 4C þ H2O þ H2

C4H4O f CO þ C3H4 (allene)

2C3H4 f C3H6 (propylene) þ H2 þ 3C

C3H4 þ C3H6 f C2H4 þ C4H6 (C4 olefins)

C3H4 þ C4H4O f C6H6 (benzene) þ CO þ H2

13

14

C4H4O þ C3H6 f C7H8 (toluene) þ H2O

C8H8 þ C3H4 f C9H8 (indene) þ C2H4

450

539

600

280

500

270

450 600

266 40

500

22

450

13

600

2

500

10

450

15

600

40

500 450

41 42

600

264

500

254

450

250

600

158

500

160

450

161

600 500

91 96

450

98

propylene, butyldiene, cyclopentadiene, hexadienyne, and hexatriene and forms toluene, styrene, indene, naphthalene, and methylindene, respectively. The reaction between furan and propylene to form toluene is found to be thermodynamically favorable (Table 9, reaction 13) and has been studied elsewhere.8 Under the entire reaction conditions used in this study, ∼55% of the C4 olefins were butadiene. In addition, two different hexatrienes were formed that had a 1:1 carbon ratio. Assume that one 625

dx.doi.org/10.1021/cs200103j |ACS Catal. 2011, 1, 611–628

ACS Catalysis

RESEARCH ARTICLE

Table 10. Experimentally Observed and Thermodynamically Calculated Carbon Molar Ratios of Various Productsa (B þ T)/(E þ P),

WHSV effect

temp effect

E/(BTXSIN),

WHSV (h1)

temp (°C)

furan conversion

1.95

600

0.97

1.17

1.09

10.16

1.28

10.34

600

0.48

1.09

1.1

3.72

1.04

21.86

600

0.28

0.83

1.43

7.74

1.73

Exp

Cal

Exp

Cal

Exp

10.34

450

0.22

1.27

537

0.86

23

1.95

0.26

0.95

10.34

500

0.32

1.28

399

0.92

40

1.65

0.23

0.85

10.34

600

0.48

1.09

247

1.1

98

3.72

0.18

1.04

10.34

650

0.6

0.83

203

1.43

143

7.74

0.16

1.73

carbon ratio

(B/T), carbon ratio

B/N, carbon ratio

molratio

Cal

a

P, propylene; E, ethylene; B, benzene; T, toluene; E, ethylene; X, xylenes; S, styrene; I, indene; N, naphthalene; Exp, experimentally observed ratio; Cal, thermodynamically calculated ratio. Furan partial pressure: 6 Torr for each run.

Table 11. Suggested Reaction Pathways of Furan Conversion over HZSM-5 at 600 °C reactions

no. DielsAlder Condensation

2C4H4O (furan) f C8H6O (benzofuran) þ H2O C4H4O þ C8H6O f C12H8O þ H2O f f f 3 3 3 f coke or heavy oxygenates

1 2

Decarbonylation of Benzofuran and Furan C8H6O f CO þ C6H6 (benzene) þ C (coke)

3

C4H4O f CO þ C3H4 (allene) = C3H4 (methylacetylene)

4 Allene (Methylacetylene) Reactions

2C3H4 f C6H8 f C3H6 (propylene) þ H2 þ 3C

5

2C3H4 f C6H8 f C5H6 (cyclopentadiene) þ H2 þ C

6

2C3H4 f C6H8 f C6H6 þ H2

7

2C3H4 f C6H8 (hexatrienes) f C6H6 (hexadienyne) þ H2 f C6H6 (benzene, by cyclization) 3C3H4 f C9H12 f C7H8 (toluene) þ C2H4 (ethylene)

8 9

C3H4 þ C5H6 f C8H10 f C6H6 (benzene) þ C2H4

10

C3H4 þ C3H6 f C6H10 f C2H4 þ C4H6 (C4 olefins)

11 FuranOlefins Reaction

C4H4O þ C3H6 f C7H10O f C7H8 (toluene) þ H2O

12

C4H4O þ C4H6 (butyldiene) f C8H10O f C8H8 (styrene) þ H2O

13

C4H4O þ C5H6 (cyclopentadiene) f C9H10O f C9H8 (indene) þ H2O

14

C4H4O þ C6H6 (hexadienyne) f C10H10O f C10H8 (naphthalene) þ H2O

15

C4H4O þ C6H8 (hexatriene) f C10H12O f C10H10 (methylindene) þ H2O

16

Alkylations C7H8 þ C3H4 f C10H12 f C8H8 (styrene) þ C2H4

17

C8H8 þ C3H4 f C11H12 f C9H8 (indene) þ C2H4

18

C9H8 þ C3H4 f C12H12 f C10H8 (naphthalene) þ C2H4

19

C6H6 þ C3H6 f C9H12 f C7H8 þ C2H4

20

C7H8 þ C3H6 f C10H14 f C8H10 (xylenes) þ C2H4

21

2C3H6 f C6H12 f C4H8 (C4 olefins) þ C2H4

22

C4H8 þ C3H6 f C7H14 f C2H4 þ C5H10 (C5 olefins) f C5H6 þ 2H2 (cyclization) C5H10 þ C3H6 f C8H16 f C2H4 þ C6H12 f C6H6 þ 3H2 (cyclization)

23 24

compound.60,77 However, in this study, the HZSM-5 deactivated very rapidly (Figure 5). Therefore, the composition of the hydrocarbon pool produced from furan is considered different from that produced from methanol. The TPO results also indicated the hydrogen-poor features on furan. For example, for methanol conversion over HZSM-5 at 450 °C, the TPO curve showed a peak representing catalytic coke at around

consuming the biomass-derived feedstock (Figure 5, stability test), the only product they produce is coke.67,76 Methanol is a hydrogen-rich feedstock (H/Ceff ratio is 2) compared with furan (H/Ceff ratio is 0.5), and thus, the lives of their hydrocarbon pools are much different from each other. The hydrocarbon pool produced during methanol conversion can last for several hours, since the feedstock, methanol, is essentially a hydrogen-rich 626

dx.doi.org/10.1021/cs200103j |ACS Catal. 2011, 1, 611–628

ACS Catalysis 510520 °C;41 however, furan conversion over HZSM-5 showed the peak at 538 and 560 °C for reactions at 400 and 600 °C, respectively. The higher oxidation temperatures imply more hydrogen-poor characteristics.

5. CONCLUSION Conversion of furan to aromatics and olefins over HZSM-5 catalyst has been studied. Temperature-programmed analysis (TGAMS and in situ FTIR) showed that at room temperature, furan formed oligomers on HZSM-5. The furan adsorption uptake was 1.73 moles of furan per mole of Al site. As the oligomers were heated, they underwent a series of reactions, including DielsAlder condensation, dehydration, decarbonylation, decarboxylation, alkylation, etc., to produce aromatics and olefins at temperatures higher than 400 °C. Graphite-type coke formed during furan conversion. Oxygen was majorly removed via dehydration, decarbonylation, and decarboxylation. A continuous flow fixed-bed reactor with a HZSM-5 packed bed inside was used to elucidate the chemistry of furan conversion. Coke rapidly formed on the catalyst surface, leading to a decrease in the overall catalytic activity. Carbon monoxide, ethylene, propylene, benzene, and toluene were final products, since their selectivity increased with increasing furan conversion. However, their carbon yield decreased rapidly with time onstream due to coke deposition. At high furan conversion (0.97), most oxygen was removed by decarbonylation (18% of 25% maximum selectivity). At 450 °C, benzofuran and coke were major products that contributed 60% carbon selectivity. At intermediate temperatures (500600 °C), the product distribution was selective to aromatics, especially to benzene and toluene. However, at 650 °C, the carbon selectivity of olefins was comparable to aromatics. Olefins were produced mainly from furan instead of benzofuran or other polyring aromatics, which were identified as important intermediates. The leaching experiment showed that furan can form polymers with Mw ∼ 104, but those polymers decomposed and formed products and coke at 400 °C, which was consistent with TGAMS. Oxygenated aromatics with conjugated double bonds were identified as compounds retained inside HZSM-5. TPO results showed that furan formed a polyring aromatic structure when adsorbed on the HZSM-5 surface. The type of coke that formed on the catalyst surface was a function of the reaction temperature, with more graphitic-like coke forming at higher reaction temperatures. The benzene/toluene and (B þ T)/(E þ P) ratios were close to 1 at all reaction conditions, which was much smaller than their thermodynamic calculated ratios. We proposed a mechanism in which the furan molecules diffuse into HZSM-5 channels and are converted to intermediates such as allene by decarbonylation and benzofuran by DielsAlder condensation. Olefins are produced via oligomerizaiton and crackings of furan, allene, and olefins themselves. Aromatics are produced via alkylation, cyclization, and reactions between olefins and furan. Ethylene is a leaving group when producing aromatics and olefins or is extracted from coke. Oxygen was removed as water, CO, and CO2 throughout the reaction. A strong pore diffusion limitation was found by Weisz modulus calculation. We concluded that the reaction was far from thermodynamic equilibrium due to space confinement imposed from HZSM-5 channels. This paper shows that zeolite catalysts can be used to convert hydrogen-poor biomass resources into aromatics and olefins.

RESEARCH ARTICLE

’ AUTHOR INFORMATION Corresponding Author

*Phone: 413-545-0276. Fax: 413-545-1647. E-mail: huber@ecs. umass.edu.

’ ACKNOWLEDGMENT This material is based upon work supported as part of the Catalysis Center for Energy Innovation, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences under Award no. DE-SC0001004. ’ REFERENCES (1) Lynd, L. R.; Wyman, C. E.; Gerngross, T. U. Biotechnol. Prog. 1999, 15, 777–793. (2) Wyman, C. E. Annu. Rev. Energ. Environ. 1999, 24, 189–226. (3) Wyman, C. E.; Dale, B. E.; Elander, R. T.; Holtzapple, M.; Ladisch, M. R.; Lee, Y. Y. Bioresour. Technol. 2005, 96, 2026–2032. (4) Singh, N. R.; Delgass, W. N.; Ribeiro, F. H.; Agrawal, R. Environ. Sci. Technol. 2010, 44, 5298–5305. (5) Huber, G. W.; Iborra, S.; Corma, A. Chem. Rev. 2006, 106, 4044–4098. (6) Lin, Y. C.; Huber, G. W. Energy Environ. Sci. 2009, 2, 68–80. (7) Carlson, T. R.; Vispute, T. R.; Huber, G. W. ChemSusChem 2008, 1, 397–400. (8) Carlson, T. R.; Cheng, Y.-T.; Jae, J.; Huber, G. W. Energy Environ. Sci. 2011, 4, 145–161. (9) Diebold, J. P.; Scahill, J. W. Abstr. Pap. Am. Chem. Soc. 1987, 193, 70-CELL. (10) Goyal, H. B.; Seal, D.; Saxena, R. C. Renewable Sustainable Energy Rev. 2008, 12, 504–517. (11) Demirbas, A. Energy Sources, Part A 2007, 29, 753–760. (12) Carlson, T. R.; Tompsett, G. A.; Conner, W. C.; Huber, G. W. Top. Catal. 2009, 52, 241–252. (13) Czernik, S.; Scahill, J.; Diebold, J. J. Sol. Energy Eng. Trans.ASME 1995, 117, 2–6. (14) Adam, J.; Antonakou, E.; Lappas, A.; Stocker, M.; Nilsen, M. H.; Bouzga, A.; Hustad, J. E.; Oye, G. Microporous Mesoporous Mater. 2006, 96, 93–101. (15) Olazar, M.; Aguado, R.; San Jose, M. J.; Bilbao, J. J. Chem. Technol. Biotechnol. 2001, 76, 469–476. (16) Fabbri, D.; Torri, C.; Baravelli, V. J. Anal. Appl. Pyrolysis 2007, 80, 24–29. (17) Fabbri, D.; Fabbri, F.; Falini, G.; Baravelli, V.; Magnani, A.; Torri, C.; Maskrot, H.; Leconte, Y. J. Anal. Appl. Pyrolysis 2008, 82, 248–254. (18) Aho, A.; Kumar, N.; Eranen, K.; Salmi, T.; Hupa, M.; Murzin, D. Y. Fuel 2008, 87, 2493–2501. (19) Corma, A.; Huber, G. W.; Sauvanaud, L.; O’Connor, P. J. Catal. 2007, 247, 307–327. (20) Chen, N. Y.; Degnan, T. F.; Koenig, L. R. CHEMTECH 1986, 16, 506–511. (21) Vispute, T. P.; Zhang, H. Y.; Sanna, A.; Xiao, R.; Huber, G. W. Science 2010, 330, 12221227. (22) Haniff, M. I.; Dao, L. H. Appl. Catal. 1988, 39, 33–47. (23) Adjaye, J. D.; Bakhshi, N. N. Fuel Process. Technol. 1995, 45, 161–183. (24) Adjaye, J. D.; Bakhshi, N. N. Fuel Process. Technol. 1995, 45, 185–202. (25) Gayubo, A. G.; Aguayo, A. T.; Atutxa, A.; Valle, B.; Bilbao, J. J. Chem. Technol. Biotechnol. 2005, 80, 1244–1251. (26) Bridgwater, A. V.; Cottam, M. L. Energy Fuels 1992, 6, 113–120. (27) Sharma, R. K.; Bakhshi, N. N. Can. J. Chem. Eng. 1993, 71, 383–391. 627

dx.doi.org/10.1021/cs200103j |ACS Catal. 2011, 1, 611–628

ACS Catalysis

RESEARCH ARTICLE

(64) Pereira, C.; Kokotailo, G. T.; Gorte, R. J. J. Phys. Chem. 1991, 95, 705–709. (65) Cox, S. D.; Stucky, G. D. J. Phys. Chem. 1991, 95, 710–720. (66) Buchanan, J. S.; Santiesteban, J. G.; Haag, W. O. J. Catal. 1996, 158, 279–287. (67) Haw, J. F.; Song, W. G.; Marcus, D. M.; Nicholas, J. B. Acc. Chem. Res. 2003, 36, 317–326. (68) Haw, J. F.; Richardson, B. R.; Oshiro, I. S.; Lazo, N. D.; Speed, J. A. J. Am. Chem. Soc. 1989, 111, 2052–2058. (69) Xu, T.; Haw, J. F. J. Am. Chem. Soc. 1994, 116, 10188–10195. (70) Sassi, A.; Wildman, M. A.; Ahn, H. J.; Prasad, P.; Nicholas, J. B.; Haw, J. F. J. Phys. Chem. B 2002, 106, 2294–2303. (71) Hansen, N.; Brueggemann, T.; Bell, A. T.; Keil, F. J. J. Phys. Chem. C 2008, 112, 15402–15411. (72) Bhan, A.; Delgass, W. N. Catal. Rev.Sci. Eng. 2008, 50, 19–151. (73) Ono, Y. Catal. Rev.Sci. Eng. 1992, 34, 179–226. (74) Bjorgen, M.; Svelle, S.; Joensen, F.; Nerlov, J.; Kolboe, S.; Bonino, F.; Palumbo, L.; Bordiga, S.; Olsbye, U. J. Catal. 2007, 249, 195–207. (75) Haw, J. F.; Marcus, D. M. Top. Catal. 2005, 34, 41–48. (76) Svelle, S.; Ronning, P. O.; Olsbye, U.; Kolboe, S. J. Catal. 2005, 234, 385–400. (77) Aguayo, A. T.; Gayubo, A. G.; Castilla, M.; Arandes, J. M.; Olazar, M.; Bilbao, J. Ind. Eng. Chem. Res. 2001, 40, 6087–6098.

(28) Horne, P. A.; Williams, P. T. Fuel 1996, 75, 1043–1050. (29) Horne, P. A.; Williams, P. T. Fuel 1996, 75, 1051–1059. (30) Horne, P. A.; Williams, P. T. Renewable Energy 1996, 7 131–144. (31) Hoang, T. Q.; Zhu, X. L.; Sooknoi, T.; Resasco, D. E.; Mallinson, R. G. J. Catal. 2010, 271, 201208. (32) Zhu, X. L.; Lobban, L. L.; Mallinson, R. G.; Resasco, D. E. J. Catal. 2010, 271, 8898. (33) Hoang, T. Q.; Zhu, X. L.; Lobban, L. L.; Resasco, D. E.; Mallinson, R. G. Catal. Commun. 2010, 11, 977981. (34) Carlson, T. R.; Jae, J.; Lin, Y. C.; Tompsett, G. A.; Huber, G. W. J. Catal. 2010, 270, 110124. (35) Lifshitz, A.; Bidani, M.; Bidani, S. J. Phys. Chem. 1986, 90, 5373–5377. (36) Fulle, D.; Dib, A.; Kiefer, J. H.; Zhang, Q.; Yao, J.; Kern, R. D. J. Phys. Chem. A 1998, 102, 7480–7486. (37) Grandmaison, J. L.; Chantal, P. D.; Kaliaguine, S. C. Fuel 1990, 69, 1058–1061. (38) Kraushaar, B.; Kompa, H.; Schrobbers, H.; Schulz-Ekloff, G. Acta Phys. Chem. 1985, 31, 581–587. (39) Carlson, T. R.; Jae, J.; Huber, G. W. ChemCatChem 2009, 1, 107–110. (40) Guisnet, M.; Magnoux, P. Appl. Catal. 1989, 54, 1–27. (41) Valle, B.; Gayubo, A. G.; Aguayo, A. T.; Olazar, M.; Bilbao, J. Energy Fuels 2010, 24, 2060–2070. (42) Spoto, G.; Geobaldo, F.; Bordiga, S.; Lamberti, C.; Scarano, D.; Zecchina, A. Top. Catal. 1999, 8, 279–292. (43) Chica, A.; Strohmaier, K.; Iglesia, E. Langmuir 2004, 20, 10982–10991. (44) Chica, A.; Strohmaier, K. G.; Iglesia, E. Appl. Catal., B 2005, 60, 223–232. (45) Hernandez, V.; Ramirez, F. J.; Zotti, G.; Navarrete, J. T. L. J. Chem. Phys. 1993, 98, 769–783. (46) Sen, S.; Bardakci, B.; Yavuz, A. G.; Gok, A. U. Eur. Polym. J. 2008, 44, 2708–2717. (47) Hoffmann, P.; Lobo, J. A. Microporous Mesoporous Mater. 2007, 106, 122–128. (48) Halasz, I.; Agarwal, M.; Marcus, B.; Cormier, W. E. Microporous Mesoporous Mater. 2005, 84, 318–331. (49) Mores, D.; Stavitski, E.; Kox, M. H. F.; Kornatowski, J.; Olsbye, U.; Weckhuysen, B. M. Chem.—Eur. J. 2008, 14, 11320–11327. (50) Bordiga, S.; Ricchiardi, G.; Spoto, G.; Scarano, D.; Carnelli, L.; Zecchina, A.; Arean, C. O. J. Chem. Soc., Faraday Trans. 1993, 89, 1843–1855. (51) Yaws, C. L. Chemical Properties Handbook; 1st ed.; McGrawHill: New York, 1998. (52) Handbook of Zeolite Science and Technology; Auerbach, S. M., Carrado, K. A., Dutta, P. K., Eds.; Marcel Dekker: New York, 2004. (53) Hansen, N.; Krishna, R.; van Baten, J. M.; Bell, A. T.; Keil, F. J. J. Phys. Chem. C 2009, 113, 235–246. (54) Petryk, M.; Leclerc, S.; Canet, D.; Fraissard, J. Catal. Today 2008, 139, 234–240. (55) Ruthven, D. M.; Kaul, B. K. Ind. Eng. Chem. Res. 1993, 32, 2053–2057. (56) Ulrich, K.; Freude, D.; Galvosas, P.; Krause, C.; Karger, J.; Caro, J.; Poladli, P.; Papp, H. Microporous and Mesoporous Mater. 2009, 120, 98103. (57) Wu, P. D.; Debebe, A.; Yi, H. M. Zeolites 1983, 3, 118–122. (58) Snurr, R. Q.; Bell, A. T.; Theodorou, D. N. J. Phys. Chem. 1994, 98, 11948–11961. (59) Song, L.; Sun, Z.; Duan, L.; Gui, H.; McDougall, G. S. Microporous Mesoporous Mater. 2007, 104, 115–128. (60) Mikkelsen, O.; Kolboe, S. Microporous Mesoporous Mater. 1999, 29, 173–184. (61) Demmin, R. A.; Gorte, R. J. J. Catal. 1984, 90, 32–39. (62) Gorte, R. J. J. Catal. 1982, 75, 164–174. (63) Vollhardt, K. P. C.; Schore, N. E. Organic Chemistry: Structure and Function, 5th ed.; W. H. Freeman: New York, 2005. 628

dx.doi.org/10.1021/cs200103j |ACS Catal. 2011, 1, 611–628